POSITIVE DISPLACEMENT PUMP REVERSE FLOW

[roker]

hi all,

where can i find evidence/proof/article/etc. for the fact that reverse flow (from discharge to suction) can not occure in positive displacement pumps when it is stopped , because of the way it is constructed and nature of operation .

regards,
roker

 

[pennpiper]

Contact the Vendor

 

[Unotec]

Look at the pump's drawings of the internals

<<A good friend will bail you out of jail, but a true friend
will be sitting beside you saying ” Damn that was fun!” - Unknown>>

 

[nickelkid]

If you have a leaky suction valve you may have reverse flow.

 

[dcasto]

there are typically check valves on the suction and discharge internal to the pump.. It would take a failure of both plus the failure of the pump discharge check valve.

 

[moltenmetal]

If a positive displacement pump didn't have leakage back through the check valves, it would behave ideally, ie. it would deliver a fixed volumetric flowrate (at a fixed operating speed) independent of downstream pressure.

How many ideal PD pumps have you ever experienced?

Check valves leak, unless installed backwards by accident. That's another corrolary of Murphy's Law you can live by...

Back leakage through two check valves of the same construction in series is definitely possible, ie. the suction and discharge check valves of the pump itself. Add a third check valve at the point of injection of whatever stream this pump is generating, and make that check valve of different construction to those in the pump, and you'l reduce that likelihood considerably.

 

[sshep]

As per dcasto, piston and diapram pumps will have both a discharge and suction check valve (in addition to any included in the piping) which must both leak through. It is common in my experience to lose flow because one of the check valves gets stuck open by a small piece of gravel or something, but a back flow will require both the inlet and outlet to fail. Still, having seen one check valve fail many times, I would have to say a backflow (two check valves fail) is not possible in the case of dirty service, although I have never had it happen.

The usual over-looked hazard with PD pumps is an uncontrolled forward flow in cases where the discharge pressure (due to upset, process cycle, etc) falls below the suction- i.e. you shut the process down and find your chemical injection tank contents has flowed into the process if you didn't isolate the pump. The manufacturer can recommend a back-pressure valve at the discharge to guard against this scenario if needed.

best wishes,
sshep

 

[psafety]

I have always been informed reverse spin is a possibility. Rarely see more than one check also.

Be especially careful of blow lines (air, N2, steam) the increased velocity can over-spin a PD. Bypass around the pump for blowing down.

 

[maddocks]

PD pumps have inherently double check valves - this is the only way they can work. To get both to fail and then to push the piston back - wow, never seen it happen. Add to that the piping on discharge should have it's own check valve, this would be a real unusual event.

 

[frankj1]

Some PD pumps back flow IE vane pumps & gear pumps. Pretty much if a pump does not have a check valve in it's design, it will back flow

If you are pumping something with solids, the check valves in diaphragm pumps can also hang up allowing back flow.

 

[Unotec]

Rotor/stator, piston and hose pumps won't flow back, they can't. can they?

<<A good friend will bail you out of jail, but a true friend
will be sitting beside you saying ” Damn that was fun!” - Unknown>>

 

[JimCasey]

There are lots of types of PD pumps.
Piston pumps/diaphragm pumps use check valves.
Gear pumps don't.
Check valves can leak. So can gear pumps-leaking thru the clearances. Plus gear pumps can be driven backwards by a reversed pressure gradient.